Navigation überspringen ([Alt] + [0])

Sitemap | english Version

>> Ressourcen > Theses > Hollosi Arno, A[..] > HTML Version > 2.3 Decisions D[..]

• Über uns
  • Mitarbeiter
  • Aktuelles
  • ... mehr über uns
  • Kontaktinformationen
• Lehre
  • Lehrveranstaltungen
  • LV-Unterlagen
  • Offene Projekte
  • NEU: offene Diplomarbeiten
  • Abschlussarbeiten
• Forschung
  • Forschungsbereiche
  • Projekte
  • Publikationen
  • Partner

• Fakultät f. Informatik
• TU Graz
• TUGonline
• Impressum

2.3 Decisions During Design

[Writing Chinese is like] dancing with brush and ink ...creating the same beauty we recognize in moving water, in foam, spary, eddies, and waves, as well as in clouds, flames and weavings of smoke in sunlight. [WH76]

A well-designed dynamic symbolic language should flow so beautifully that it will create comparable effects. [Len97]

When creating a new visual language a lot of decisions have to be made. These initial decision are crucial for the ``look and feel'' of the final language. However most of the time we do not have enough knowledge to decide on a rationale basis and have to rely on our intuition and feelings instead. Let us have a look at some of the considerations and their consequences.

• Users: when designing a language one has to consider whom this language is aimed at. It is obvious that a language designed for technicians might not be suitable for children and vice versa.

• Primary objective: the primary objective of a language has a major impact on the final design. The primary objective is what the language is designed for or what is important for the design. For example, Minspeak's (see Section 2.4) primary objective is using as few ``keystrokes'' as possible to create a sentence. Vedo Vidi's (see Section 2.7) primary objective is communication across language barriers.

• Ontology: an ontology describes what ideas, words, or processes can be expressed within the language; that is, what part of the world is part of the language? For example, mathematicians use a language of symbols whose ontology is limited to mathematical expressions - it is great for its intended use, but could not be used to tell bedtime stories to children. Vedo Vidi's authors say their language has a ``postcard ontology'', meaning that it is suitable for a special way of communication, namely that used on postcards, which usually includes information about travel, family members, weather, and health.

• Structure: the structure describes how expressions are built and what they consist of. Does the language have nouns, verbs, adjectives? How are these elements arranged? Are expressions related to each other, or are they ordered?

• Dynamic versus static: are the elements of the language static, i.e. can it be printed on paper without loss of information? Or is information conveyed in a dynamic form, e.g. like a movie? Dynamically presented languages need a dynamic medium, such as computer, video, or audio. While those languages allow a rich and sophisticated set of features, they call for great care during the design and implementation of actual language frameworks.

• Extensions: languages usually have a predefined set of elements. However those elements might not cover every area desired, and therefore users may want to create new elements and include them into the language. It is important that the design of a language covers whether a language can be extended. If yes, it is crucial to define how new elements are added, because visual, computer supported languages store a lot of internal information to provide more sophisticated features. This internal information either has to be generated automatically or entered by the user, if the user wants to extend the language.

• Interaction: interaction describes the different ways a user can access information. For example, written natural language has rather limited interaction possibilities. The recipient can read the text, browse over it, use an index, or use the table of contents. Languages designed for use on computers may feature more interaction. Using again the example of written natural language, many sophisticated functions can be added on a computer: searching, hyper-links, agents that look for a similar topic, a background library with in-depth explanations of words, different versions of the text, annotations, comparisons, and much more. The user is no longer limited to a rather static representation, but can use the story as gateway to explore the world beyond the story.

• Easy authoring: a language may have many features for the user (or in other words: reader), but it should be easy to create (write) too. Of course, sophisticated tools can help in creating messages or automate some processes, but if the structure of the language is too complicated writing can be a tiresome process. It might not be easy to find a good compromise between reading abilities and writing needs.